Review Article Big Data Management for Cloud-Enabled Geological Information Services

Page created by Lois Adkins
 
CONTINUE READING
Review Article Big Data Management for Cloud-Enabled Geological Information Services
Hindawi
Scientific Programming
Volume 2018, Article ID 1327214, 13 pages
https://doi.org/10.1155/2018/1327214

Review Article
Big Data Management for Cloud-Enabled Geological
Information Services

          Yueqin Zhu ,1,2 Yongjie Tan ,1,2 Xiong Luo ,3,4 and Zhijie He                                   3,4

          1
            Development and Research Center, China Geological Survey, Beijing 100037, China
          2
            Key Laboratory of Geological Information Technology, Ministry of Land and Resources, Beijing 100037, China
          3
            School of Computer and Communication Engineering, University of Science and Technology Beijing (USTB), Beijing 100083, China
          4
            Beijing Key Laboratory of Knowledge Engineering for Materials Science, Beijing 100083, China

          Correspondence should be addressed to Yueqin Zhu; yueqin zhu@126.com and Xiong Luo; xluo@ustb.edu.cn

          Received 20 October 2017; Revised 10 December 2017; Accepted 31 December 2017; Published 29 January 2018

          Academic Editor: Anfeng Liu

          Copyright © 2018 Yueqin Zhu et al. This is an open access article distributed under the Creative Commons Attribution License,
          which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

          Cloud computing as a powerful technology of performing massive-scale and complex computing plays an important role in
          implementing geological information services. In the era of big data, data are being collected at an unprecedented scale. Therefore,
          to ensure successful data processing and analysis in cloud-enabled geological information services (CEGIS), we must address the
          challenging and time-demanding task of big data processing. This review starts by elaborating the system architecture and the
          requirements for big data management. This is followed by the analysis of the application requirements and technical challenges
          of big data management for CEGIS in China. This review also presents the application development opportunities and technical
          trends of big data management in CEGIS, including collection and preprocessing, storage and management, analysis and mining,
          parallel computing based cloud platform, and technology applications.

1. Introduction                                                          information to knowledge, and from knowledge to wisdom.
                                                                         In addition, it provides data analysis, mining, organization,
In the era of big data, the data-driven modeling method                  and management services for the scientific management of
enables us to exploit the potential of massive amount of                 land resources, prospecting breakthrough strategic action
geological data easily [1–3]. In particular, by mining the               and social services, while conducting multilevel, multiangle,
data scientifically, one can offer new services that bring               and multiobjective demonstration applications on geological
higher values to customers. Furthermore, it is now possible              data for government decision-making, scientific research,
to implement the transition from digital geology to intelli-             and public services [5].
gent geology by integrating multiple systems in geological                   Big data technologies are bringing unprecedented oppor-
research through the use of big data and other technologies              tunities and challenges to various application areas, especially
[4].                                                                     to geological information processing [2, 6, 7]. Under these
     The application of geological cloud makes it possible               circumstances, there are some advancements achieved in the
to fully utilize structured and unstructured data, including             development of this area [8, 9]. Furthermore, from various
geology, minerals, geophysics, geochemistry, remote sens-                disciplines of science and engineering, there has been a
ing, terrain, topography, vegetation, architecture, hydrology,           growing interest in this research field related to geological
disasters, and other digitally geological data distributed in            data generated in the geological information services (GIS).
every place on the surface of the earth [4, 5]. Moreover,                We analyze the number of those documents indexed in “Web
the geological cloud will enable the integration of data                 of Science” [10]. In Figures 1 and 2, we can easily find that, in
collection, resource integration, data transmission, informa-            the past ten years, the number of those documents in which
tion extraction, and knowledge mining, which will pave                   “geological data” is in the “Title” and in the “Topic” are both
the way for the transition from data to information, from                increasing, respectively. Hence, the analysis for geological
Review Article Big Data Management for Cloud-Enabled Geological Information Services
2                                                                                                               Scientific Programming

              250                                                        different fields, because the nature and types of data vary in
              200                                                        different fields and in different problems. Geology is a data
                                                                         intensive science and geological data are characterized with
              150
     Number

                                                                         multisource heterogeneity, spatiotemporal variation, correla-
              100
                                                                         tion, uncertainty, fuzziness, and nonlinearity. Therefore, the
                                                                         geological cloud has a certain degree of confidentiality and
               50                                                        it is highly domain-specific; meanwhile, it is developed on
                                                                         the basis of a large amount of geological data accumulated
                0
                     2007 2008 2009 2010 2011 2012 2013 2014 2015 2016   over a long period of time [5, 11]. There are many real-time
                                            Year                         data generated from some issues like geological disasters and
                                                                         geological environment. The geological cloud includes core
                      The number of documents                            basic data, which can be divided into three parts: existing
Figure 1: The trend of the number of documents in which “geolog-         structured database, some unstructured data, and public
ical data” is in the “Title” from 2007 to 2016.                          application data. Therefore, it is important to take good
                                                                         advantage of the existing traditional structured data, use the
                                                                         big data technologies to deal with the relevant unstructured
              6000
                                                                         data, and also consider the peripheral public data.
              5000                                                            Geological big data are multidimensional, and they
              4000                                                       consist of both structured and unstructured data [12]. The
    Number

                                                                         technical methods of big data analysis differ greatly from
              3000
                                                                         those of professional databases. Long-term geological survey,
              2000                                                       geological study, and years of geological information accu-
              1000                                                       mulation have formed a rich and professional database, which
                 0
                                                                         is an important fundamental assurance for land and resources
                     2007 2008 2009 2010 2011 2012 2013 2014 2015 2016   science management, geological survey, and geological infor-
                                            Year                         mation public service [13]. This “professional cloud” objec-
                                                                         tively requires the technology research and development,
                       The number of documents                           such as construction of professional local area network,
Figure 2: The trend of the number of documents in which                  data sharing platform, and geological big data visualization
“geological data” is in the “Topic” from 2007 to 2016.                   services. Hence, the construction of geological cloud is closely
                                                                         related to land resource management, deployment decision,
                                                                         and the application demand of public service. The key tech-
                                                                         nologies of research and development include the following:
data in GIS is an interesting and important research topic               unstructured data extraction and mining analysis, structured
currently.                                                               and unstructured data mixed storage and management, big
    Considering the development status of cloud-enabled                  data sharing platform, data transmission, and visualization
geological information services (CEGIS) and the applica-                 [11].
tion requirements of big data management analysis, this                       Generally, the construction of geological cloud is a long-
article describes the significant impact and revolution on               term systematic project. This means that it is required to
GIS brought by the advancement of big data technologies.                 follow the basic principles of “standing on the reality, focusing
Furthermore, this article outlines the future application                on the future” and “focusing on the long-term and overall
development and technology development trend of big data                 situation, embarking on the current and local situation,” in
management analysis in CEGIS.                                            order to achieve the analysis and application of geological
    The remainder of this article is organized as follows.               cloud public data and core data gradually in accordance with
In Section 2, we provide a review on CEGIS, with an                      the technical route of big data analysis; thus the construction
emphasis on the descriptions for the system architecture and             of geological cloud will be implemented eventually. For the
those requirements from big data management. Then, the                   earth, the land and resources management should cover
challenges for big data management in CEGIS are presented                many respects, including human behavior, climate change,
in Section 3. Furthermore, the key technologies and trends               development and utilization of various resources, natural
on big data management in CEGIS are analyzed in Section 4.               disasters, environmental pollution, and the ecosystem cycle.
Finally, conclusion is drawn in Section 5.                               Then, the introduction of big data technologies can integrate
                                                                         this type of resource information to provide the ability of
2. Review on Cloud-Enabled Geological                                    uniformly dealing with the problems related to the entire
   Information Services                                                  earth information resources, which has a significant effect on
                                                                         the strategic planning of land and resources management [3].
The construction of geological cloud differs from the current                 The geological cloud is an important part in the science
big data analysis based on Internet and Internet of Things               system for geological data research. The ultimate goal for
(IoT). Having a deep understanding of data characteristics is            developing geological cloud is to better describe and under-
necessary to collect, process, analyze, and interpret data in            stand the complex earth system and geological framework,
Review Article Big Data Management for Cloud-Enabled Geological Information Services
Scientific Programming                                                                                                                        3

                                                               Geological cloud
                                                                                       Data         Computing
                      Business
                      network                                                       resources        resources
                                                              Geological                                          Internet
                      private                                 information                                        public area
                      area                                    service
  Communication                                                                                                                Satellite remote
     satellite
                                                                                                                                    sensing

                            Field data collection

                                                                                        Data analysis
                                                                                         processing
                          Marine
                     geological survey

                                                        Data services

                                           Figure 3: The system architecture of geological cloud.

provide the scientific basis for the description of the land                      including data exchange and audit, can be carried out
surface and the biodiversity characteristics of the earth, and                    by single-directional light gate.
improve the ability to deal with complex social problems.
                                                                            (iii) “One Main Node and Three Domain-Specific Nodes.”
                                                                                  One main node is constructed in China Geologi-
2.1. System Architecture. Because the business service func-                      cal Survey Development Research Center. In addi-
tions of each country are different, the system architecture of                   tion, three domain-specific nodes—namely, marine
the geological cloud would vary. In the following, we present                     node, geological environment node, and aviation geo-
a system architecture in Figure 3 [14], using China as an                         physical exploration and remote sensing node—are
example.                                                                          constructed, respectively. Each node is configured
     The geological cloud combines the geological survey                          with the corresponding servers, storage equipment,
Intranet and the geological survey Extranet. It enables the                       network equipment, management platform, large-
sharing and management of computing resources, storage                            scale specialized data processing software system and
resources, network resources, software resources, and geolog-                     various customized applications. Each node would
ical data resources [15].                                                         store huge amounts of geological data and conform
     Geological cloud can be summarized with the following                        to current data security standards. The master node
characteristics [14].                                                             and the domain-specific nodes are linked via optical
                                                                                  fibers. The master node will consist of 200 computing
    (i) “One Platform: The Geological Cloud Management
                                                                                  nodes with 3 PB storage capacity and will be equipped
        Platform.” It uniformly manages computing
                                                                                  with some geological data processing software system.
        resources, storage resources, network resources,
                                                                                  The master node will be hosted in a medium-sized
        software resources, and geological data resources.
                                                                                  supercomputing center and it will provide support for
   (ii) “Two Networks: The Geological Survey Intranet and                         the three-dimensional seismic exploration data pro-
        the Geological Survey Extranet.” Here, the Intranet is                    cessing and other large-scale computing. The three
        constructed by creating a network that is physically                      domain-specific nodes are to maintain their scale in
        isolated from the Internet. The Intranet is developed                     the near future to facilitate reasonable scheduling and
        on the basis of the existing geological survey network                    efficient utilization of information resources and data
        and each node is linked through a dedicated line or                       resources.
        bare fiber. All of the internal business management
        systems, software systems, and data are deployed                    On the Extranet, it deploys a system for geological
        on the Internet, providing services to 28 local units           survey business management and auxiliary decision-making.
        and those users of more than 350 geological survey              The system provides a real-time tracking and management
        projects. Facilitated by the public geological survey           function for geological survey projects and various resources.
        network, the geological survey business management                  Main users of the geological cloud include institutional
        system, geological data information service system,             users, geological survey project users, and the general public
        and public geological data can be deployed on the               users. The institutional users can store the current geological
        Extranet accessed by the general public. The com-               database and newly collected data in the geological cloud
        munication between the Intranet and the Extranet,               through the geological survey business network and can
Review Article Big Data Management for Cloud-Enabled Geological Information Services
4                                                                                                             Scientific Programming

                                              Cloud-Enabled Geological Information Service

                                                         Big Data Management

                                           The Large-scale Parallel Processing Framework Using
                                                               MapReduce

                                                               Typical Tools

                                                                                          PowerView
                                      Sqoop                                              Karmasphere

                                                       HDFS                 Hibe
                                                       Hbase               Mahout

                                                   1               2                3                  4
                                      Data Pre-          Data               Data             Result
                                      Processing        Storage        Mining/Analysis       Display

                                          Figure 4: Schematic diagram of big data analysis.

obtain the geological data of other institutions from the cloud         and service, through the use of advanced cloud computing
as needed. The geological survey project users can access               system, IoT, and big data processing flow. It is shown in
the cloud geological background data through 4G or satellite            Figure 4 [5].
lines and can collect data through data the collection system.
                                                                        3. Challenges for Big Data
2.2. Requirement from Big Data Management. The construc-                   Management in Cloud-Enabled
tion of geological cloud must meet customer demand. Big                    Geological Information Services
data technologies are then used as the means to implement
the geological cloud.                                                   Geological big data are generated regarding various layers of
    The types and quantity of geological data have been                 the earth, the history of the conformation and evolution of
continuously growing over the years. Geological data include            the earth, and the material composition of the earth and its
all kinds of electronic documents, structured, semistruc-               changes. It also involves the exploration and utilization of
tured, and unstructured data, such as various databases (map            mineral resources. In the current geological work, the collec-
database, spatial database, and attribute database), pictures,          tion, mining, processing, analysis, and utilization of various
tables, video, and audio. Generally speaking, those important           complex type data are closely related to those general big data.
data may be buried in the massive data without the guidance             The “4V” characteristics of big data—namely, Volume, Veloc-
for requirements. Hence, the first step is to understand the            ity, Variety, and Veracity—also apply to geological big data.
user requirements and then gain the capability of large-scale
data processing. This is followed by data mining, algorithm,            3.1. Volume. Currently, there is no consensus on the size of
and analysis, which will ultimately generate value. Big data            geological data. Geological big data are a collection of data,
technologies in the field of geography must meet different              including geology, minerals, remote sensing, geophysical
needs from people at different levels, including the public             exploration, geochemical exploration, surveying, and map-
demand of the geologic data services and professional data              ping, which are interconnected and integrated. In terms of the
demand for geological research institutions, as well as related         number of mines, there are at least 70000 in China, and some
enterprises and government departments [16].                            official documents and popular science books indicate that
    On the basis of big data analysis technologies, a complete          there are more than 200000 deposits and minerals that have
data link is formed connecting data, information, knowledge,            been found. The information is huge and cannot be processed
Review Article Big Data Management for Cloud-Enabled Geological Information Services
Scientific Programming                                                                                                             5

using conventional tools. For example, an Excel spreadsheet       database (covering the national exploration right, mining
cannot contain all the information of 70000 mining areas.         right, mining right verification, geological information meta-
Then, it is difficult to classify and rank the 200000 mines, so   data database, and many others) [17].
it is necessary to rely on the concepts and technologies of big       For those databases, they are still expanding and con-
data [17].                                                        summating, and their practical values have not yet been
     Especially in recent years, images, video, and other types   fully reflected. However, the vast majority of researchers are
of data have emerged on a large scale. With the application       virtually impossible to have all of the above data, at most,
of 3D scanning and other devices, the data volume has             using their own accumulated data. Anyway, even if their
been increasing exponentially. The ability to describe the        accumulated data, both on the quantity and on type, is incom-
data is more and more powerful, and the data are gradually        parable by 10 years and 20 years ago, they are, in fact, in the era
approximated to the real world. In addition, the large amount     of the “relatively big data.” From 1999 to 2004, for example,
of data is also reflected in the aspect that the methods and      in “the Chinese mineralization system and regional metalor-
ideas used by people to deal with data have undergone             ganic evaluation” project, although there are 202 national
a fundamental change. In the early days, people used the          academic experts that participated in it, they only master
sampling method to process and analyze data in order to           data of 4500 properties (all kinds of minerals). From 2006 to
approximate the objective with a small number of subsample        2013, the study of “national important mineral and regional
data. With the development of technologies, the number of         mineralization laws” was conducted; meanwhile, the mining
samples gradually approaches the overall data. Using all the      area covered only by the mineral resources research institute
data can lead to a higher accuracy, which can explain things      was 30600. Therefore, the increase of information and the
in more detail [18].                                              amount of data are unprecedented in the last ten years.
     Recently, the China geological survey system has built
databases including regional geological database (cover-
                                                                  3.2. Variety. From the formal point of view, the geological big
ing the 1 : 2500000, 1 : 1000000, 1 : 500000, 1 : 250000, and
                                                                  data have many characteristics, including multidimensional-
1 : 200000 regional geological map; the national 1 : 200000
                                                                  ity, multiscale, and multitenses. And they contain structured,
natural sand; the isotope geological dating; and the lithos-
                                                                  semistructured, and unstructured data and usually are stored
tratigraphic unit database), basic geological database (cov-
                                                                  in forms of text, graphics, images, databases (including image
ering the national rock property database and national geo-
                                                                  database, spatial database, and attribute database), tables,
logical working degree database), mineral resources database
                                                                  videos, and audios in a fragmented state. For example, a
(covering the national mineral resources, the national min-
                                                                  large number of field outcrop description data, borehole
eral resources utilization survey mining resources reserves
                                                                  core description data, and all kinds of geological survey,
verification results, the national survey of large and medium-
                                                                  exploration report, and a large number of geological maps,
sized mines, the prospect of mineral resources, the survey of
                                                                  drawings, and photos were stored and managed in the form
the resources potential of major solid mineral resources in
                                                                  of paper for a long time; even the numerous relational
China, and the geological and mineral resources database),
                                                                  databases and spatial databases were primarily used to store
oil and gas energy database (covering the oil and gas basins
                                                                  and manage structured data that are tabulated and vectorized,
in China, the geological survey results of the national oil
                                                                  while the text descriptions, records, and summaries were
and gas resources, the national petroleum and geophysical
                                                                  directly stored. Very few standardized processing and struc-
exploration, national shale gas, national coal bed methane,
                                                                  tural transformations were performed. Furthermore, there is
national natural gas hydrate, and other databases), geophys-
                                                                  no tool available to effectively integrate storage and manage
ical database (covering 1 : 1 million, 1 : 500000, 1 : 250000,
                                                                  structured, semistructured, and unstructured data.
1 : 200000, and 1 : 50000 gravity, national regional grav-
ity, national aeromagnetism, national ground magnetism,
national electrical survey, seismic survey, national aviation     3.3. Velocity. The increase of geological data is very fast,
radioactivity, and national logging database), geochemi-          especially in remote sensing geology, aviation geophysical
cal database (covering the databases of national 1 : 250000       exploration, regional geochemical exploration, and other
and 1 : 20 geochemical exploration, national multiobjective       fields, due to the introduction of new technologies and
geochemical and national land quality evaluation results),        new methods. Meanwhile, high speed processing is also a
remote sensing survey database (covering national aeronau-        characteristic of big data. In addition to the need of analyzing
tical remote sensing image, China resources satellite data,       data in real time, people also need to describe the results
space remote sensing image, national mine environmental           of data mining and processing through the use of several
remote sensing monitoring, national high score satellite, and     data processing techniques, such as image and video, while
other databases), drilling database (covering the national        requiring effective and efficient handling skills. For example,
geological borehole information, the national important           the detection of the deep earth information not only needs
geological borehole, the Chinese mainland scientific drilling     to obtain parameters of the seismic wave reflection and
core scanning image library, and so on), hydraulic cycle          refraction but also needs to conduct quick processing, so as
hazards database, data literature database, special subject       to timely predict whether earthquake will occur and forecast
database (covering the national mineral resources poten-          the time, location, and intensity. In this way, we can avoid
tial evaluation database, the important mineral “three-rate”      the disaster effectively. When applying a variety of data to
investigation and evaluation database), work management           a particular mountain, one should learn which ones have
Review Article Big Data Management for Cloud-Enabled Geological Information Services
6                                                                                                          Scientific Programming

spatial limitations and which are not related to spatiality,         4. Key Technologies and Trends on Big Data
so that one deduces the metallogenic law and guides the                 Management in Cloud-Enabled Geological
prospecting better [17].                                                Information Service (CEGIS)
3.4. Veracity. For the understanding of the value of big data,       With the rapid advancement of big data technologies, some
most people consider it low value density. It means that the         key technologies are accordingly developed for big data
real useful information in the vast amount of data is very little.   management in CEGIS. Specifically, a schematic diagram of
Taking video as an example, the useful data may be only a            those key technologies is shown in Figure 5. Then, in this
second or two in the continuous monitoring process. While            section we present an analysis on those key technologies.
big data is high value, it does not need to be invested too          Meanwhile, the trends along this direction are also discussed.
much; just collecting information from the Internet can bring
business value. Therefore, big data has the characteristics          4.1. Geological Big Data Collection and Preprocessing. Geolog-
of low value density and high business value. The same is            ical big data collection and preprocessing aim to categorize
true for geological big data. So far, there has been a lot of        those geological big data obtained from geological data,
information about geophysical prospecting, but only a few            geological information, and geological literature.
have been confirmed, and the discovered mines were less. But
once a breakthrough was made, its socioeconomic value was
enormous, such as the lithium polymetallic deposit in Tibet          4.1.1. Geological Data Collection Access. In addition to the
and the newly discovered Jima copper polymetallic deposit in         traditional collection ways, it is also required to carry out
the outskirts of Sichuan [17].                                       large-scale network information access and provide real-
    In addition, the spatial attribute and temporal attribute of     time, high concurrency, and fast web content acquisition,
geological data also bring a big challenge to data accuracy.         combining with the application characteristics in the cloud
Any geological data have spatial attribute, and their values         environment. Currently, considering that the growth rate of
are reflected in the spatial law of distribution of mineral          geological information generated from the network is very
resources. For this reason, in the process of establishing           fast, the big data analysis system should obtain relevant data
the metallogenic series, exploring the metallogenic law, and         quickly.
constructing the mathematical model, the spatial attribute
of the metallogenic model should be considered. Obviously,           4.1.2. Quality and Usability Characteristics of Geological Data.
every metallogenic series has its spatial attribute. Geological      It needs to distinguish and identify valuable information
data also has the time attribute, which is very different from       through intelligent discovery and management technologies.
physical, chemical, and other natural sciences. One of the           Because the information value density contained in different
fundamental pillars of geology is the geological time scale.         data sources differs from each other, filtering out the useless
The rocks, strata, and deposits of different geological periods      or low-value data source can effectively reduce the data stor-
have different distribution characteristics and regularity, so       age and processing costs. Then, it can also further improve
those data have their own time attribute.                            the efficiency and accuracy of analysis.
    It is obvious that those characteristics of geological big
data mentioned above impose very challenging obstacles               4.1.3. Geological Data Entity Recognition Model. According
to the data management in CEGIS. The challenges related              to the subject domain of geology, the distributed data are
to geological big data management can be summarized as               extracted to form a data warehouse, after conducting the
follows:                                                             operation of processing and integration. When extracting
     (i) It is quite difficult to describe and model geological      data in the field of geology, it needs to use entity modeling
         big data, since there are few effective characteris-        method to abstract entities from the vast numbers of data,
         tics description mechanisms and object modeling             so as to find out the relationship between those entities. This
         approaches under the cloud computing environment.           approach ensures that the data used in warehouse data can
                                                                     be consistent and relevant in accordance with the data model
    (ii) There remain many technical issues that must be
                                                                     [19]. These recognized data are directly input into the system,
         addressed to fully manage, mine, analyze, integrate,
                                                                     stored as metadata, which could be used for data management
         and share those geological big data, in consideration
                                                                     and analysis.
         of those complex characteristics, including multi-
         source heterogeneous data, highly spatiotemporal
         variation, high-volume and high-correlation data,           4.1.4. Aggregation of Geological Big Data. Generally, dif-
         and many others.                                            ferent data sources and even the same data source may
    (iii) Many issues appear in achieving decision support,          generate data with different formats. As mentioned above,
          such as data incompleteness, data uncertainty, and         because these structural, semistructured, and unstructured
          high-dimensionality of data.                               multimodal geological big data are integrated together, the
                                                                     data heterogeneity is obvious in big data analysis. Then,
    The broad range of challenges described here make good           data aggregation as the key technology in achieving data
topics for research within the field of big data management in       extraction and transformation [20] enables data sharing and
CEGIS. They are analyzed in the next section.                        data fusion between heterogeneous data sources. Through
Review Article Big Data Management for Cloud-Enabled Geological Information Services
Scientific Programming                                                                                                              7

  Multi-Source Heterogeneous Data
  Multimodal Data
  Highly Spatio-Temporal Variation
  High-Volume and High-Correlation Data
  Low-Value-Density Data
  Complex and Uncertain Data

                                                                            Applications
          Characteristics: “4V”
    (Volume, Variety, Velocity, Veracity)

                                                                         Analysis and Mining
                                            Big Data Flow

                                                                                                                 Key Technologies
             Geological Big Data
                                                                               Storage

                                                                            Preprocessing

                                                                             Collections

                                                                     Highly Performable Big Data
                                                                      Cloud Computing Platform

                                                                        Big Data Management

                             Figure 5: Schematic diagram of key technologies for big data management in CEGIS.

the use of heterogeneous information aggregation technolo-                 Here, we provide an example of aggregating and collect-
gies, the unified data retrieval and data presentation could           ing geological big data in CEGIS. Figure 6 illustrates this
be achieved. On the basis of it, after aggregating those               process. While developing CEGIS, all kinds of geological
distributed heterogeneous data sources, they are extracted             data should be processed. Through the use of geological
and converted to achieve the functions of automatically                cloud, big data are collected, and then they are aggregated to
constructing subject domain database and data warehouse                achieve some key functions in geological information service
[21].                                                                  platform, including catalog sharing, intelligent searching,
                                                                       data products release, and collaborative service.
4.1.5. Management of Geological Big Data Evolution Tracking
Records. In order to effectively utilize geological big data, it       4.2. Geological Big Data Storage and Management. From
needs to track the evolution of big data during the whole life         the data collection perspective, geological data can be
cycle of GIS, with the purpose of achieving the traceable big          divided into field survey data, drilling and engineering explo-
data management.                                                       ration data, remote detection data, analytical test data, and
Review Article Big Data Management for Cloud-Enabled Geological Information Services
8                                                                                                                      Scientific Programming

                                                                                 Organizing by rules
                                                                                           Regional
                                                                                                                            Geological
                                                                                           geology
                                                                                                                             Cloud
                                                                                           Energy
                                                                                           geology
           Geological information service platform
                                                                                           Mineral                        Intelligent data
     Catalog    Intelligent   Data products Collaborative                                  geology                           collecting
     sharing     searching      release        service                                    Hydrogeology
                                                               Sharing and                                                Location based
                                                                                                                              service
                                                               Exchanging
                                                               by
                                                               Permissions        Aggregating by levels                  Background data
                                                 Aggregating                                              Collecting         support
                                                  big data                                                 big data
                                                                                                                         Data uploading to
                                                                                       Programme                               cloud

                                                                                                                          Online analysis
                                                                                       Project team
                                                                                                                                         Intelligent
                                                                                                                                         survey in
                                                                                                                                         the
                                                                                                                                         terminal
                                                                                    Field work group

                              Figure 6: An example of aggregating and collecting geological big data in CEGIS.

comprehensive study data. From the angle of comprehensive                    a hybrid distributed storage model through the use of cloud
application fields, they can be also divided into regionally                 advantages of flexibly scalable deployment, to meet the users’
geological survey data, energy and mineral resources evalu-                  requirement for geological big data resource management
ation and exploration survey results data, geological disaster               with satisfactory data durability and high availability [23].
monitoring and early warning data, geological environment                        Here, the hot research topics include the following:
survey and evaluation results data, and marine geological
                                                                                 (i) For geological applications, the load optimization
survey and evaluation data. From the data formality point
                                                                                     storage should be implemented to achieve the cou-
of view, they can be divided into picture data, text report                          pling for data storage and application and the cou-
data, tabular data, and image data. These data are collected                         pling for distributed file system and the new storage
by different units.                                                                  system.
    Facing these complex geological big data mentioned
above, the traditional relational database will be difficult to                 (ii) Based on the application characteristics of distributed
                                                                                     databases, more studies could be conducted on the
handle them, while the distributed storage system can be used
                                                                                     application of new databases NoSQL and NewSQL in
to store such huge amounts of data and manage them. Then,
                                                                                     geological survey work.
the data system places the massive data in many machines,
which avoids such limitation of storage capacity, and also                       With the development of big data technologies, more and
brings many problems that have not occurred before in stand-                 more mature distributed data storage solutions will emerge
alone systems. Hence, some distributed data storage solutions                and will be applied to big data analysis [24, 25].
have accordingly emerged, including Hadoop, Spark, and                           Specifically, in the management of geological big data,
other nonrelational database systems (like HBase, MongoDB,                   the implementation of data query—for example, spatial
and many others) [22]. These different solutions satisfy                     query—has been a long-term focus. Generally, considering
the specific requirements from different applications. When                  those advantages with unified modeling language (UML) and
applying to the analysis of big data, different solutions can                computer-aided software engineering (CASE) methodology,
be employed according to the specific needs of different                     the spatial database could be accordingly designed and
intelligence analysis. Furthermore, different solutions can be               implemented to characterize and realize the object-oriented
combined to meet specific needs. Actually, there have been                   spatial vector big data firstly [26]. And then, in the developed
some attempts to develop combination strategies for dis-                     spatial database, the function of self-generating codes would
tributed storage model, varying in the big data management                   be achieved to realize two-way spatial query between graphic-
performance requirement, and the complexity of collected                     objects and property data [26]. Moreover, in consideration of
big data that are supported by the distributed storage system                the complex characteristics of geological big data, the spatial
[23]. Hence, there is still a room for improvement and                       query is achieved finally through the use of Flex technology in
optimization of geological big data storage, while designing                 ArcGIS Server software platform [27]. Practically speaking,
Review Article Big Data Management for Cloud-Enabled Geological Information Services
Scientific Programming                                                                                                             9

in this technology, the spatial query could be implemented           deal with terabytes or even petabytes of data, as well as
through two functions, including “Query” and “Find” query            multidimensional data, all kinds of noise data and dynamic
methods [28].                                                        data. Because data mining algorithms will directly influence
                                                                     the outcome of the discovered knowledge, selecting the most
4.3. Geological Big Data Analysis and Mining. In terms of geo-       appropriate algorithms and parallel computing strategy is the
logical data analysis and mining, it needs to combine geolog-        key to data mining.
ical data, geological information, and geological literatures,           Effective data mining also could reduce manual inter-
through the analysis of geological application demand of             vention during information processing and make use of
real-time mining, to explore geological big data environment         methods and tools of big data intelligent analysis [35, 36].
analysis and mining algorithm, in an effort to fully achieve         Recently, there has been a growing interest in the geological
the goal of intelligent mining for geological big data.              big data mining through the use of some novel computational
     Figure 7 shows a schematic diagram of discovering geo-          intelligent methods—for example, rough set [37] and fuzzy
logical knowledge through analyzing and mining geological            aggregation [38]. Moreover, with the development of those
big data. It can be easily found that geological big data analysis   neural network based machine learning algorithms in recent
and mining play an important role in achieving the final goal.       years, some popular methods, including extreme learning
More relevant research work related to it mainly involves the        machine [39, 40], approximate dynamic programming [41],
following aspects.                                                   and kernel learning [42], could be used to further improve
                                                                     mining effectiveness for geological big data in the future.
4.3.1. Geological Big Data Analysis. Considering the special
applications, geological big data technologies would apply big       4.4. Highly Performable Big Data Cloud Computing Platform.
data concepts to analyze the metallogenic rules by making            Highly performable big data cloud computing platform is the
full use of various data related to ore, to recognize deposit        foundation for big data analysis. It enables parallel computing
metallogenic series, to summarize the metallogenic regulari-         for large-scale incremental real-time data and large-scale
ties and express in an appropriate way (like voice, image, and       heterogeneous data [43–46].
many others), and to establish the scientifically mathematical            With the advent of massive data storage solution,
model. The model then uses new exploration data to predict           many big data distributed computing frameworks have
future data and to guide geological prospecting.                     been proposed. Among them, Hadoop, MapReduce, Spark,
     In addition, it is necessary to pay special attention to        and Storm are the most important distributed computing
the analysis of new geological big data information collected        frameworks. These frameworks have different characteristics
from social medium and networks [29]. These include the              and solve different problems in applications [47–50]. The
geological text information flow data from microblog web             Hadoop/MapReduce is often used for offline complex big
sites, the geological multimedia data from media sharing             data processing, the Spark is often employed in offline fast
web sites, the geology-related user interaction data on social       big data processing, and the Storm is often available for
networking web sites, and many others [30]. These mul-               real-time online big data processing. Different computing
tisource data complement traditional big data. Specifically,         frameworks have their different advantages and disadvan-
such data should be addressed with the help of multilingual          tages. Hadoop/MapReduce is easy to program, and it is with
information processing, multilingual machine translation,            satisfactory scalability and fault tolerance. In addition, it is
and social network cross-language retrieval [31]. Big data           suitable for offline processing of massive data with petabyte
analysis of such data is a key to deep use of geological data in     level, but it does not support real-time computation and flow
a broader dimension. With the maturity of big data analysis          calculation. Spark is a memory-based iterative computing
technologies, it becomes possible to analyze and extract             framework. By placing intermediate data in memory, Spark
valuable information from these data [32] and to provide             can achieve higher iterative calculation performance. The
effective solutions for geological big data applications.            programming model of Spark is more flexible than that
                                                                     of Hadoop/MapReduce, but Spark is not suitable for those
4.3.2. Geological Big Data Mining. Data mining is to extract         applications in which the fine-grained updates are conducted
the unknown and useful knowledge and information from                asynchronously. Hence, Spark may be unavailable for those
the massive multilevel spatiotemporal data and attribute data,       application models that require incremental changes. Storm
using statistics, pattern recognition, artificial intelligence,      is suitable for stream data processing. It can be used to handle
set theory, fuzzy mathematics, cloud computing, machine              a stream of incoming messages and can write the processed
learning, visualization, and relevant techniques and methods.        result to a specified storage device. Another major application
Data mining could reveal the relationship and evolution              of Storm is real-time data processing where data are not
trend behind the geological big data, achieve the automatic          necessary to be written into storage devices, which usually
or semiautomatic acquisition of the new knowledge, and               results in low time delay. Hence, Storm is particularly suitable
provide the decision basis for resource prediction, prospect-        for scenarios where real-time online analysis is required to
ing, environmental assessment, and disaster prevention and           obtain results for big data analysis.
mitigation [33]. Therefore, the knowledge is obtained directly            An application example is geological big data aggregation
from known geological data to provide relevant decision              mining framework based on Hadoop [16]. Geological big
support [34]. In consideration of the amount of data, it may         data aggregation mining platform research is based on the
Review Article Big Data Management for Cloud-Enabled Geological Information Services
10                                                                                                                                     Scientific Programming

                                               Filtering               Preprocessing                                  Converting
                       Original
                                                data                      data                                          data
                       geological
                       data

                                                    Evaluating                                        Analyzing and
                      Geological                      results                                         mining data
                      knowledge

                                                             Practicability of machine learning and statistical algorithms
                                                             Feasibility of intelligent mining
                                                             Effectiveness of online analysis and mining
                                                             ···

          Figure 7: Schematic diagram of discovering geological knowledge through analyzing and mining geological big data.

China geological survey data network, and it uses the Hadoop                                                          The 20th exploratory line

                                                                          Element content
                                                                                            100
technology to improve and modify existing platform, to make
it suitable for big data applications, and to provide a platform                            50
for the pilot applications. The geological survey grid platform
can be updated in three layers—that is, the virtual layer,                                    0
                                                                                                  1     4   7 10 13 16 19 22 25 28 31 34 37 40
the computing layer, and the terminal application layer. The
virtual layer is the virtualization of computer resources based                                         The element of Mn
on Hadoop distributed file system (HDFS) virtualization                                                 The element of Co
technology, which is the foundation of cloud computing and                                              The element of Be
cloud services. The computing layer mainly uses MapReduce
                                                                                        Figure 8: Correlation among three element morphologies.
method to implement the analysis algorithms for geological
big data. Currently, the geological big data technologies
mainly use the block calculation strategy to achieve parallel
analysis through the utilization of the characteristics of
                                                                       the metallogenic law, it is necessary to fully understand the
Hadoop, in an effort to speed up the analysis and processing
                                                                       massive data about spatial distribution, reserves and produc-
of geological data. The terminal application layer is designed
                                                                       tion in mineral origin, the geological structure of the mineral
to display the results and receive user feedback to improve
                                                                       origin, and related geological survey data. Then, it is to
system availability.
                                                                       conduct the regular speculation and objectivity expression of
     MapReduce has been used to perform morphological cor-
                                                                       these geological big data, so that one can identify the essential
relation analysis, which involves the analysis of geochemical
                                                                       reasons for the distribution of mineral origin. Actually, using
data processing and the study of the correlation between mul-
                                                                       geological big data technologies could help to translate data
tielements. Figure 8 shows the pattern correlation between
                                                                       into new understanding or knowledge and help to guide the
elements. It can be seen from Figure 8 that the elements of
                                                                       future geological prospecting work.
Mn, Co, and Be are similar in the distribution of morphology.
Therefore, from a qualitative point of view, the correlation
is relatively high. Moreover, after testing, the proposed              4.5.2. Smart Prospecting. The types of deposits vary, and
prototype system is running three times more quickly than              the formation of them is related to certain geological back-
the existing common computing platform, showing that the               grounds and geological effects, respectively. The geological
geological big data is applicable to the Hadoop platform.              backgrounds include tectonic unit and stratigraphic unit,
Furthermore, some applications of using MapReduce could                deep upper mantle and lithosphere conditions, and paleo-
be found in [51].                                                      geography and palaeoclimate environment on the surface of
                                                                       the earth. Geological effects include tectonism, magmatism,
4.5. Applications of Geological Big Data Technologies                  sedimentation, metamorphism, and weathering. Theses geo-
                                                                       logic backgrounds and effects in the wide range of space,
4.5.1. Exploration of Metallogenic Law. The metallogenic law           and in the long geologic history, are a dynamic change
is the human regular knowledge of the temporal and spatial             and repeated stack, and large deposits can be formed only
distribution of mineral resources, and its cognitive level,            in a variety of favourable conditions. Long-term scien-
ability, and scope are all related to the size of data, the type       tific research and experience accumulation formed mineral
of data, and the way of data processing. Therefore, to deduce          deposit and mineralization prediction subject. Professionals
Scientific Programming                                                                                                              11

are guided by certain theories and methods to adopt quanti-         knowledge of the entity. Due to the continuous release of
tative or qualitative methods to predict prospecting with the       user-generated content and linking open data on the Internet,
existing knowledge and experience.                                  people need to explore knowledge interconnection methods
    However, in view of the difficulties of geological data shar-   which both conform to the development of the network
ing and the limitations of calculation tools and calculation        information resources and meet users’ requirements from
methods, most of the known deposits in the past are indepen-        a new perspective according to the knowledge organization
dent of each other. In the future, we can use geological data       principles in the large data environment, to reveal human
to connect several adjacent deposit exploration data, conduct       cognition on a deeper level [55].
unified analysis and specialized processing, determine the               In this context, knowledge graph (KG) was formally put
“digital” characteristics of the distribution of metallogenic       forward by Google in May 2012, and its goal is to improve the
materials, find out metallogenic potential, delineate the           search results and describe the various entities and concepts
abnormal area and prospective area, and promote geological          that exist in the real world and the relationship between these
prospecting. Furthermore, geological data informatization           entities and concepts. KG is a great choice to select the essence
and standardization could be improved [52].                         and discard the dross, as well as the sublimation of the present
                                                                    semantic web technology. In recent years, the applications of
                                                                    KG have been increasing rapidly, and there is now a mature
4.5.3. Service of People’s Livelihood Geology. After entering
                                                                    method used to draw a KG and conduct intelligent searching
the 21st century, geological work is more closely related
                                                                    research based on KG [34]. However, the function of KG
to economic development, and geological work plays an
                                                                    has not been fully implemented at present, especially for the
important role in every aspect of social and economic life.
                                                                    specific object of geological big data; the application aspect
Agricultural geology, urban geology, environmental geology,
                                                                    still needs to be further strengthened. Along this direction,
tourism geology, disaster geology, and other works have been
                                                                    the visualization service for geological data in the web-based
strengthened, and the service area has also been expanded
                                                                    system is attracting more and more attention [56, 57].
[53]. Meanwhile, the public demand for geological informa-
tion is increasingly urgent [54].
    In order to meet the social demand for geological data,         5. Conclusion
China Geological Survey carried out the construction of
                                                                    Big data technologies make it possible to process massive
geological cloud, which built cluster geologic data service
                                                                    amount of unstructured and semistructured geological data.
system with the National Geological Information Center
                                                                    And the geological cloud enables us to explore the application
and the Provincial Geological Information Center as the
                                                                    of demand-driven geological core data and to extract new
backbone nodes, conducted the integration of data resources,
                                                                    information from unstructured data, while supporting the
and applied the GIS cloud technology, in order to obtain
                                                                    decision-making in land resources management. Thus, the
large-scale computing ability and solve those key problems,
                                                                    geological cloud could effectively organize and use geological
such as the distributed storage, processing, query, interop-
                                                                    big data, to mine the data scientifically, with the purpose
erability, and virtualization of massive spatial data [5, 13].
                                                                    of producing higher value and achieving the corresponding
Recently, in China, Shandong Provincial Bureau of Geology
                                                                    service.
and Mineral Resources also carried out the construction of
                                                                         In the architecture of geological cloud, this article
“the application system of geological business based on e-
                                                                    describes the application background of CEGIS and the
government cloud platform.” It mainly relies on the public
                                                                    demands from big data management. Furthermore, we elab-
service cloud platform of the e-government in Shandong
                                                                    orate the application requirements and challenges faced in
province and constructs the government external network
                                                                    big data management technologies. Then, more analyses are
service system and Internet service system to achieve the
                                                                    provided from four aspects, including data size, data type,
unified management and information service of the mineral
                                                                    data processing speed, and data processing accuracy, respec-
resource. Using technical methods of spatial analysis, big
                                                                    tively. In addition, this article outlines the research status and
data mining, and three-dimensional geological model, it
                                                                    technology development opportunities of big data related in
develops a basic system framework for geological mining
                                                                    CEGIS, from the perspectives of big data acquisition and pre-
services, featured by “a (cloud) platform, a (data) center,
                                                                    processing, big data storage and management, big data analy-
and many application systems,” to improve the ability of the
                                                                    sis and mining, highly performable big data cloud computing
people’s livelihood geological service, promote interaction
                                                                    platform, and big data technology applications. With the con-
with the public, realize socialization services, and promote
                                                                    tinuous development of big data technologies in addressing
the clustering and industrialization of the mineral resources
                                                                    those challenges related to geological big data, such as the
information services.
                                                                    difficulties of describing and modeling geological big data
                                                                    with some complex characteristics, CEGIS will move towards
4.5.4. Application of Knowledge Visualization Service. With         a more mature and more intelligent direction in the future.
the continuous development of web technology, human
beings have experienced the “Web 1.0” era, which is charac-         Conflicts of Interest
terized by document interconnection, and “Web 2.0”, which
is characterized by data interconnection, and are moving            The authors declare that there are no conflicts of interest
towards the new “Web 3.0” era based on the interconnected           regarding the publication of this article.
12                                                                                                                     Scientific Programming

Acknowledgments                                                           [17] D. Wang, X. Liu, and L. Liu, “Characteristics of big geodata
                                                                                and its application to study of minerogenetic regularity and
This work was supported in part by the Key Laboratory of                        minerogenetic series,” Mineral Deposits, vol. 34, no. 6, pp. 1143–
Geological Information Technology of Ministry of Land and                       1154, 2015.
Resources under Grant 2017320, the National Key Technolo-                 [18] B. Pan and R. Yang, “Management and utilization of big data
gies R&D Program of China under Grant 2015BAK38B01,                             for geology,” Surveying and Mapping of Geology and Mineral
and the National Key R&D Program of China under Grant                           Resources, vol. 33, no. 1, pp. 1–3, 2017.
2016YFC0600510.                                                           [19] P. Yang and L. J. Lu, “The research on encoding methodology
                                                                                of the character of geological entity based on mass geological
                                                                                data,” Advanced Materials Research, vol. 962-965, pp. 208–212,
References                                                                      2014.
 [1] P. Vermeesch and E. Garzanti, “Making geological sense of ‘big       [20] X. Luo, D. Zhang, L. T. Yang, J. Liu, X. Chang, and H. Ning, “A
     data’ in sedimentary provenance analysis,” Chemical Geology,               kernel machine-based secure data sensing and fusion scheme in
     vol. 409, pp. 20–27, 2015.                                                 wireless sensor networks for the cyber-physical systems,” Future
                                                                                Generation Computer Systems, vol. 61, pp. 85–96, 2016.
 [2] J. Chen, J. Xiang, Q. Hu et al., “Quantitative geoscience and
     geological big data development: a review,” Acta Geologica           [21] C.-L. Kuo and J.-H. Hong, “Interoperable cross-domain seman-
     Sinica, vol. 90, no. 4, pp. 1490–1515, 2016.                               tic and geospatial framework for automatic change detection,”
                                                                                Computers & Geosciences, vol. 86, pp. 109–119, 2016.
 [3] Y. Zhu, Y. Tan, R. Li, and X. Luo, “Cyber-Physical-Social-
     Thinking modeling and computing for geological information           [22] Y.-J. Wang, W.-D. Sun, S. Zhou, X.-Q. Pei, and X.-Y. Li,
     service system,” International Journal of Distributed Sensor               “Key technologies of distributed storage for cloud computing,”
     Networks, vol. 12, no. 11, 2016.                                           Journal of Software , vol. 23, no. 4, pp. 962–986, 2012.
                                                                          [23] M. Armbrust, A. Fox, R. Griffith et al., “Above the clouds: a
 [4] M. M. Song, Z. Li, B. Zhou, and C. L. Li, “Cloud computing
                                                                                berkeley view of cloud computing,” Tech. Rep. UCB/EECS-
     model for big geological data processing,” Applied Mechanics
                                                                                2009-28, University of California at Berkeley, California, Calif,
     and Materials, vol. 475-476, pp. 306–311, 2014.
                                                                                USA, 2009.
 [5] J. P. Chen, J. Li, N. Cui, and P. P. Yu, “The construction and
                                                                          [24] J. Xia, Z. Bai, B. Wang, J. Chang, and Y. Wu, “Design and
     application of geological cloud under the big data background,”
                                                                                implementation of comprehensive management platform for
     Geological Bulletin of China, vol. 34, no. 7, pp. 1260–1265, 2015.
                                                                                geological data informatization,” Acta Scientiarum Naturalium
 [6] C. Li, “The technical infrastructure of geological survey infor-           Universitatis Pekinensis, vol. 50, no. 2, pp. 295–300, 2014.
     mation grid,” in Proceedings of the 18th International Conference
                                                                          [25] W. Hua, J. Liu, and X. Liu, “Data management of object
     on Geoinformatics, pp. 1–6, 2010.
                                                                                type geological features on control dictionary,” Earth Science -
 [7] L. Wu, L. Xue, C. Li et al., “A geospatial information grid                Journal of China University of Geosciences, vol. 40, no. 3, pp. 425–
     framework for geological survey,” PLoS ONE, vol. 10, no. 12,               430, 2015.
     Article ID e0145312, 2015.
                                                                          [26] B. Jia, C. Wang, C. Liu, and W. W. Sun, “Design and implemen-
 [8] K. Evangelidis, K. Ntouros, S. Makridis, and C. Papatheodorou,             tation of object-oriented spatial database of coalfield geological
     “Geospatial services in the Cloud,” Computers and Geosciences,             hazards -Based on object-oriented data model,” in Proceedings
     vol. 63, no. 2, pp. 116–122, 2014.                                         of the 2010 International Conference on Computer Application
 [9] M. Huang, A. Liu, T. Wang, and C. Huang, “Green data gath-                 and System Modeling, ICCASM ’10, pp. V1282–V1286, Taiyuan,
     ering under delay differentiated services constraint for inter-            China, IEEE, October 2010.
     net of things,” Wireless Communications and Mobile Comput-           [27] http://server.arcgis.com/.
     ing, 2018, http://downloads.hindawi.com/journals/wcmc/aip/
                                                                          [28] X. Zhou, X. Li, A. Chen et al., “Design and implementation of
     9715428.pdf.
                                                                                the service system of spatial data for geological data,” Journal of
[10] https://www.webofknowledge.com/.                                           Geomatics, vol. 38, no. 4, pp. 57–60, 2013 (Chinese).
[11] C. Yang, M. Yu, F. Hu, Y. Jiang, and Y. Li, “Utilizing Cloud Com-    [29] H. Huang, Z. Chao, and C. Feng, “Opportunities and challenges
     puting to address big geospatial data challenges,” Computers,              of big data intelligence analysis,” CAAI Transactions on Intelli-
     Environment and Urban Systems, vol. 61, pp. 120–128, 2017.                 gent Systems, vol. 11, no. 6, pp. 719–727, 2016.
[12] L. Wu, L. Xue, C. Li et al., “A knowledge-driven geospatially        [30] S. Jin, W. Lin, H. Yin, S. Yang, A. Li, and B. Deng, “Community
     enabled framework for geological big data,” ISPRS International            structure mining in big data social media networks with
     Journal of Geo-Information, vol. 6, no. 6, article no. 166, 2017.          MapReduce,” Cluster Computing, vol. 18, no. 3, pp. 999–1010,
[13] Y. Tan, “Architecture and key issues of geological big data and            2015.
     information service project,” Geomatics World, vol. 23, no. 1, pp.    [31] C. C. Yang, C.-P. Wei, and L.-F. Chien, “Managing and mining
     1–9, 2016.                                                                 multilingual documents: Introduction to the special topic issue
[14] Y. Tan, “Architecture investigation of the construction of geo-            of information processing management,” Information Process-
     logical big data system,” Geological Survey of China, vol. 3, no.          ing & Management, vol. 47, no. 5, pp. 633-634, 2011.
     3, pp. 1–6, 2016.                                                    [32] X. Luo, J. Deng, W. Wang, J.-H. Wang, and W. Zhao, “A
[15] W. He and Y. Wang, “Prototype system of geological cloud                   quantized kernel learning algorithm using a minimum kernel
     computing,” Progress in Geophysics, vol. 29, no. 6, pp. 2886–              risk-sensitive loss criterion and bilateral gradient technique,”
     2896, 2014.                                                                Entropy, vol. 19, no. 7, article 365, 2017.
[16] Y. Zhu, Y. Tan, J. Zhang, B. Mao, J. Shen, and C. Ji, “A             [33] C. H. Tse, Y. L. Li, and E. Y. Lam, “Geological applications
     framework of hadoop based geology big data fusion and mining               of machine learning in hyperspectral remote sensing data,” in
     technologies,” Acta Geodaetica et Cartographica Sinica, vol. 44,           Proceedings of Conference on Image Processing - Machine Vision
     no. S0, pp. 152–159, 2015.                                                 Applications VIII, 2015.
You can also read